Google Data Analytics Professional Certificate Data Analysis with R Programming Syllabus

What is a tibble and how does it differ from a standard data frame?

A tibble is the tidyverse's streamlined data frame: it never changes input data types or variable names, never creates row names, and prints only the first 10 rows for easier viewing.

How does R represent missing values, and how can you handle them?

Missing values appear as NA. Functions like is.na() detect them, many functions accept na.rm = TRUE to ignore them, and drop_na() removes rows containing them.

What is a function in R? Give an example.

A body of reusable code that performs a specific task, called by name with arguments in parentheses — e.g. mean(c(2, 4, 6)) returns $\frac{2+4+6}{3} = 4$, and print("hello") displays text.

What are arguments in an R function?

The information (values, variables, or datasets) supplied inside a function's parentheses that the function needs to run, e.g. in sqrt(16), the argument is 16 and the result is $\sqrt{16} = 4$.

What does the pipe operator %>% do in R?

It passes the result of one function directly into the next, chaining operations in readable order — e.g. data %>% filter(x > 5) %>% arrange(x) means 'take data, then filter, then arrange'. It replaces deeply nested function calls.

What is the difference between nested functions and pipes in R?

Nested functions place one call inside another and are evaluated from the inside out (e.g. arrange(filter(data, x > 5), x)), which becomes hard to read; pipes (%>%) express the same steps top-to-bottom in the order they happen.

What is an R package, and where do most packages come from?

A package is a shareable unit of reusable code, documentation, sample data, and tests that extends R. Most are downloaded from CRAN (the Comprehensive R Archive Network), which vets submissions.

What is the difference between install.packages() and library()?

install.packages("name") downloads and installs a package once onto your machine; library(name) loads an installed package into the current session so you can use it — this must be done every new session.

What is the tidyverse and which core packages does it include?

A collection of R packages sharing a common design for data analysis. Core packages: ggplot2 (visualization), dplyr (manipulation), tidyr (tidying), readr (importing), plus tibble, purrr, stringr, and forcats.

What are the roles of dplyr, tidyr, readr, and ggplot2 in the tidyverse?

dplyr — data manipulation (filter, select, mutate, arrange, summarize); tidyr — reshaping/cleaning data into tidy format; readr — importing rectangular data such as CSV files; ggplot2 — creating data visualizations.

How do you import a CSV file into R with the tidyverse?

Use readr's read_csv("filename.csv"), which reads the file into a tibble, e.g. df <- read_csv("sales.csv"). Base R's equivalent is read.csv().

Which functions give you a quick preview or summary of a data frame in R?

head() shows the first six rows, View() opens a spreadsheet-like viewer, str() and glimpse() show structure/column types, colnames() lists column names, and summary() gives summary statistics.

What does the data() function do in R?

It lists and loads the preloaded example datasets that come with R and its packages (e.g. data("diamonds") loads the diamonds dataset from ggplot2) for practice and testing.

What does the clean_names() function from the janitor package do?

It automatically converts column names to a consistent format containing only letters, numbers, and underscores (e.g. 'First Name' becomes first_name), ensuring unique, syntactically valid names.

In dplyr, what do filter() and select() each do?

filter() keeps rows that meet a logical condition (e.g. filter(price > 100)); select() keeps or drops columns by name (e.g. select(name, price)). Filter works on rows; select works on columns.

What does the mutate() function do in R?

It adds new columns to a data frame or modifies existing ones without removing others, e.g. mutate(weight_kg = weight_lb / 2.2) creates a new calculated column.

What do the separate() and unite() functions do?

separate() splits one column into multiple columns using a separator (e.g. splitting 'full_name' into first and last); unite() is the opposite, combining multiple columns into one.

What do rename() and rename_with() do in dplyr?

rename() changes individual column names (new_name = old_name); rename_with() applies a function to many names at once, e.g. rename_with(df, tolower) makes all column names lowercase.

How do arrange(), group_by(), and summarize() work together to organize data?

arrange() sorts rows (ascending by default; use desc() for descending); group_by() groups rows by one or more columns; summarize() then computes aggregate statistics per group, e.g. summarize(mean_price = mean(price)).

What is the purpose of the bias() function in R and which package provides it?

From the SimDesign package, bias() compares actual outcomes with predicted outcomes to quantify bias in a model — an average difference near zero suggests minimal bias.

What is ggplot2 based on conceptually, and what is its core idea?

It implements the 'grammar of graphics': every plot is built from a dataset, aesthetics (aes — visual properties like x, y, color, size mapped to variables), and geoms (geometric shapes like points, bars, lines that represent the data).

Write the basic ggplot2 code pattern for a scatterplot and explain each part.

ggplot(data = df) + geom_point(mapping = aes(x = var1, y = var2)). ggplot() creates the coordinate system from the data, geom_point() draws points, and aes() maps variables to the x and y axes. Layers are added with +.

Match common ggplot2 geoms to the chart types they create.

geom_point() — scatterplot; geom_bar() — bar chart (counts rows by default); geom_line() — line chart; geom_smooth() — trend/smoothed line; geom_histogram() — histogram; geom_boxplot() — box plot.

What is faceting in ggplot2, and which two functions provide it?

Faceting splits one plot into multiple small subplots, one per subgroup of the data. facet_wrap(~variable) facets by a single variable; facet_grid(rows ~ cols) facets by two variables in a grid.

How do you add titles, subtitles, captions, and text notes to a ggplot2 chart?

labs() adds labels outside the plot grid — title, subtitle, caption, and axis labels; annotate() places text or shapes at specific coordinates inside the plot, e.g. annotate("text", x = 3, y = 50, label = "Peak").

How do you save a plot created with ggplot2?

Use ggsave("filename.png"), which by default saves the last plot displayed, with the file type inferred from the extension (e.g. .png, .pdf); you can also use the Export option in RStudio's Plots pane.

What is R Markdown and why is it useful for analysts?

A file format (.Rmd) for making dynamic documents that combine narrative text (in Markdown), executable R code chunks, and their output in one place — letting analysts document and share a fully reproducible analysis.

What are the parts of an R Markdown document?

A YAML header at the top (between --- lines, holding title, author, date, and output format), Markdown-formatted text sections, and code chunks delimited by three backticks with {r}, which can be run individually.

What does 'knitting' an R Markdown document mean, and to which formats can you export?

Knitting (the Knit button, powered by the knitr package) executes all code chunks and renders the .Rmd file into a shareable report — commonly HTML, PDF, or Word; other outputs include slide presentations and dashboards.

How does an R notebook differ from a plain R Markdown document, and how can notebooks be shared?

An R notebook is an R Markdown document that lets you run code chunks interactively and displays output directly beneath each chunk, saving an .nb.html preview alongside the .Rmd. Notebooks are shared as rendered HTML/PDF/Word files or published on platforms like RPubs, GitHub, or Kaggle.